Retinoic acid (RA) induces growth-arrest and differentiation of murine S91 melanoma cells, and serves as an excellent model for malignant transformation of melanocytes. S91 cells express RAR alpha, beta, and gamma, and all three RAR-isoforms can induce growth-arrest. However, only activation of RARg by the RARg-selective agonist CD437 induces differentiation. CD437, unlike RA, also triggers substantial concomitant apoptosis before the remaining cells become differentiated. CD437 and RA may differentially potentiate RARg-mediated transcription, and the promoters of primary target genes of RARg may harbor new classes of RAREs and/or allow specific receptor/cofactor interactions in the context of these elements. Specific residues in RARg are responsible for agonist-dependent DNA-binding activity and protein/protein interactions. Initial studies will clone CD437 and RA-regulated primary target genes, and then clone their promoters and identify RAREs. Regulation by RARg and RARg-cofactor interactions on these elements will be studied. The investigator will also test whether CD437-regulated genes induce apoptosis. Finally, the investigator will perform mutational analysis of RARg in order to identify the domains involved in the interaction with the response elements and/or cofactors. The result from these studies will allow the integration of the 3-D structure of RARg with its molecular and biological actions in relation to an important disease.